Pith. sign in

REVIEW 4 major objections 5 minor 50 references

Sparis: Neural Implicit Surface Reconstruction of Indoor Scenes from Sparse Views

T0 review · 4 major / 5 minor · reviewed 2026-08-10 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗ Review JSON

Pith's one-line read Sparis reconstructs indoor surfaces from 10–20 images by replacing monocular depth with triangulated inter-image matches.

desk verdict Solid engineering extension of VolSDF with matching-based depth priors; the headline gains are real-looking but the evaluation needs tightening and the depth-prior claim is under-measured. read the letter →

arxiv 2501.01196 v1 pith:Z5QHIECL submitted 2025-01-02 cs.CV

Yulun Wu , Han Huang , Wenyuan Zhang , Chao Deng , Ge Gao , Ming Gu , Yu-Shen Liu This is my paper · ORCID
classification cs.CV
keywords sparse-viewreconstructionindoorsceneneuralimplicitsurfacesigneddistancefunctioninter-imagematchingdepthpriorcross-viewreprojection3D
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

The paper claims that sparse-view indoor reconstruction collapses because monocular depth priors cannot be scaled correctly from only a few views. It replaces monocular depth with an inter-image matching prior: matched pixel pairs across views are triangulated with known camera poses to give absolute depth anchors, and a reprojection loss ties the rendered surface back to the matched pixels. Two filters, an angular score for choosing source views and an epipolar weight for downweighting geometrically inconsistent matches, make the prior robust to matching errors. On ScanNet with 15–20 views the method reaches an F-score of 0.647 versus 0.464 for NeuRIS, and on Replica with 10 views 0.825 versus 0.454 for MonoSDF. If correct, practical room-scale 3D reconstruction from a handful of photos becomes feasible with neural implicit surfaces.

What carries the argument

The load-bearing object is the inter-image matching prior built from a pretrained dense feature matching network. For each image pair the network returns pixel correspondences and a confidence; with known camera poses these are triangulated into absolute depth values that supervise the neural SDF renderer. The two supporting mechanisms are an angular filter, a certainty-weighted angular score used to select the source view with favorable triangulation geometry, and an epipolar weight function that downweights matches by their Sampson distance to the epipolar constraint, as expressed in equations (12)–(15). Together they convert raw matching output into depth and reprojection supervision that is less sensitive to matching noise.

What would settle it

Take a sparse indoor sequence with large textureless regions and wide-baseline image pairs, run the matching network, and compare the density and accuracy of the triangulated depth anchors against ground-truth depth. If match coverage is too low or the Sampson-weighted correspondences still drift, the reported F-score advantage over monocular-prior baselines would shrink or disappear. A controlled ablation that swaps in a weaker matching network while keeping all other training fixed should degrade reconstruction quality in proportion to match quality.

Watch

Extended reading notes

Core claim

The central claim is that inter-image correspondence, not monocular depth, is the right geometric prior for sparse-view indoor surface reconstruction. The authors argue that monocular depth supervision requires estimating a global scale and shift, which is ill-posed when overlap between views is small, so the geometry collapses. Instead they extract dense pixel matches between image pairs, triangulate those matches into 3D points and hence absolute depth, and supervise the signed-distance-field renderer with an inter-image depth loss. A cross-view reprojection loss then forces the rendered surface point along a ray to reproject onto the matched pixel in the other view, enforcing consistency across views. The angular filter and epipolar weight reduce the influence of wrong or weakly constrained matches. With these pieces, the method reportedly produces smoother, more complete and more accurate meshes than prior indoor reconstruction methods under the same sparse-view settings.

Load-bearing premise

The load-bearing assumption is that the pretrained feature matching network supplies enough correct, dense, geometrically consistent correspondences across sparse indoor views, especially on textureless walls and ceilings and over wide baselines; if matches are too sparse or noisy, triangulated depth anchors and reprojection constraints lose accuracy and the two filters cannot repair the missing matches. The main comparisons also assume known camera poses.

Editorial extensions

If this is right

  • Indoor surface reconstruction no longer requires hundreds of views: with 10–20 images the method reports F-scores of 0.647 on ScanNet and 0.825 on Replica, ahead of the monocular-prior baselines it compares against.
  • Monocular depth, even when scaled optimally by least squares, is the point of failure in the sparse regime; replacing it with absolute triangulated depth removes the scale ambiguity.
  • Cross-view reprojection consistency acts as a regularizer that reduces overfitting when view overlap is low.
  • The angular filter and epipolar weight make the reconstruction resilient to matching noise, and this robustness carries over to estimated camera poses: with COLMAP poses the method still reports 0.514 F-score, above 0.464 for NeuRIS with ground-truth poses.
  • The same priors transfer to object-level sparse reconstruction, giving Chamfer distance comparable to the leading object-level method on DTU with 3 views.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • A natural extension the paper leaves implicit is refining the matching network or fusing it into the training loop, since improvements in match quality should directly improve triangulated depth and reprojection.
  • The COLMAP-pose experiment implies pose error costs roughly 0.13 F-score; joint refinement of poses and geometry could recover part of that gap.
  • Scenes with repetitive texture or wide baselines will stress the pairwise matcher, so the practical operating envelope is set by match density on textureless surfaces, which the paper does not quantify.
  • The inter-image depth and reprojection scheme could be inserted into other neural field or Gaussian-splatting renderers, not only the SDF-based pipeline used here.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes Sparis, a neural implicit surface reconstruction method for indoor scenes from sparse views. Instead of relying on monocular depth priors, it uses inter-image feature matching: RoMa correspondences between view pairs are triangulated into absolute depth targets, which supervise a VolSDF-based neural surface rendering; a reprojection loss encourages cross-view consistency; and an angular filter plus an epipolar weight suppress unreliable matches. The method is evaluated on ScanNet (15-20 views) and Replica (10 views), where it reports large F-score improvements over NeuRIS, MonoSDF, and other baselines, and ablations show each component contributes. The supplementary provides per-scene tables, additional comparisons with Gaussian-splatting methods and DUSt3R, and a discussion of pose assumptions.

Significance. If the results are reliable, Sparis offers a meaningful step for sparse-view indoor reconstruction by replacing scale-ambiguous monocular depth priors with absolute depths triangulated from learned correspondences. The idea is simple and uses an off-the-shelf matcher, and the paper includes useful ablations and per-scene supplementary tables. The central claim, however, is weakened by evaluation choices: a key baseline is retuned from its default, failed scenes are excluded from averages, no error bars or seed variance are reported, and the claimed accuracy of the inter-image depth prior is never directly measured. The method is not conceptually circular, but the empirical support for its main causal claim is incomplete.

major comments (4)
  1. [Comparison, Tables 1-2] The baseline comparisons are not even-handed. MonoSDF is changed from its default monocular depth weight (0.1) to 0.001 because the default 'unable to produce valid meshes,' and NeuS and HelixSurf averages are computed only over scenes where they produced valid meshes (4 and 1 failures, respectively, out of 10 ScanNet scenes). This selective reporting biases the headline improvements (0.647 vs 0.464 on ScanNet, 0.825 vs 0.454 on Replica). Please report per-scene results for all methods, count failed scenes as F-score 0 or as a separate 'failed' category, and provide default-vs-retuned MonoSDF numbers so readers can assess the effect of the retuning.
  2. [Inter-Image Depth Loss, Eqs. (7)-(15)] The central claim that inter-image matching provides 'more accurate depth information' is not directly tested. The paper never reports the number of RoMa matches per view pair, the distribution of Sampson distances, or the error of the triangulated depths eD(r) relative to ground-truth depth on the evaluation scenes. On textureless walls, floors, and ceilings, match density could be low and the angular filter and epipolar weight cannot invent missing correspondences. Without diagnostics on match density and triangulation accuracy, the large F-score gains cannot be attributed to the proposed depth prior rather than to the normal prior or to the reprojection loss acting as regularization. Please add such quantitative analysis for ScanNet and Replica.
  3. [Experiments, Table 5 and Section D of Supplementary] The main comparisons in Tables 1 and 2 use ground-truth poses for all methods, while COLMAP poses are evaluated only for the proposed method (Table 5). Since the triangulated depth prior is directly sensitive to pose accuracy, the comparison is not symmetric. Please run MonoSDF, NeuRIS, HelixSurf, and the proposed method with the same COLMAP poses and report the results; if those baselines are too pose-sensitive to run, state that limitation explicitly and justify why the GT-pose comparison is the relevant one.
  4. [Ablation Study and Table 3; Tables 1-2] No measure of variability is reported. With only 10 ScanNet scenes and 8 Replica scenes, and with baselines retuned or selectively averaged, the reported gaps could be within training stochasticity. Please report mean and standard deviation over multiple seeds for the proposed method and the main baselines, or otherwise justify that the results are stable. This is especially important because the neural rendering training and the sampling-based losses are stochastic.
minor comments (5)
  1. [Eq. (14)] The term '1ui r,s' appears to be a typo; it should be '(1 − ui r,s)' as in Eq. (7).
  2. [Section 'Experiments and Anaysis'] 'Anaysis' is a typo and should be 'Analysis'.
  3. [Ablation Study] The list of ablation settings repeats the numbering '(4)' twice; the settings should be numbered 1 through 5.
  4. [Supplementary Table 4] The footnote for DUSt3R says 'GT poses are included as inputs,' but DUSt3R is designed to operate without poses; please clarify what was provided and why this is a fair comparison.
  5. [Implementation Details] For reproducibility, please specify the exact number of views used for each ScanNet scene, the camera intrinsics used, and whether a code release is planned.

Circularity Check

0 steps flagged · score 1.0 of 10

No significant circularity: depth and reprojection priors come from independent pretrained matching and poses, not from the predicted surface.

full rationale

The derivation chain is not circular. In Eq. (6) the matching pairs (pa, pb, u) come from the pretrained RoMa network f_phi applied to input images; Eq. (7)/(14) uses the triangulated depth eD as a fixed target for the rendered depth D_hat, and Eq. (9)/(15) uses the matched pixel p_s as a fixed target for the reprojected coordinate p_s'. Both targets are computed from the input images, camera poses, and pretrained weights before and independently of the SDF optimization; they are not functions of the predicted surface. The angular filter (Eqs. 10-11) and epipolar weight (Eqs. 12-13) gate these external matches rather than define the output. The normal prior is likewise an external Omnidata prediction, and the ablations (Tables 3-4) demonstrate the components' contributions against external baselines on ScanNet and Replica. The only self-citation, NeuSurf (Huang et al. 2024b), is used as a comparison baseline and as the DTU evaluation protocol; it is not load-bearing for the method's central claim. Concerns that RoMa match density and triangulated-depth error are not directly reported are robustness or evidence gaps, not circularity. Retuning MonoSDF or excluding failed scenes affects comparison fairness, not the logical dependence of the claimed prediction on its inputs.

Assumptions & free parameters 4 free parameters · 5 assumptions · 0 invented entities

The central claim rests on external pretrained models (RoMa for matching, Omnidata for normals), known camera poses, and hand-set loss weights. There are no invented physical entities. The method is an optimization scheme whose success is empirically asserted rather than formally proven.

free parameters (4)
  • loss weights lambda_1, lambda_2, lambda_3, lambda_4 = 0.01, 0.01, 0.05, 0.05
    Hand-chosen hyperparameters in Eq. 16, tuned on the evaluation benchmarks; no sensitivity analysis or cross-validation reported.
  • gamma (sigmoid scale) in epipolar weight = 0.1
    Hand-chosen scaling in Eq. 12; the paper reports no sensitivity analysis for this value.
  • epsilon in angular filter threshold = 0.001
    Hand-chosen threshold in Eq. 11 for source view selection; no sensitivity analysis reported.
  • Softplus beta in network = 100
    Fixed activation scale in supplement Eq. 1; part of the architecture, not central to the claim.
assumptions (5)
  • domain assumption Camera poses are known and sufficiently accurate
    The method triangulates matches using poses (Eqs. 6-8); the main tables use GT poses. The supplement shows robustness with COLMAP poses only for the proposed method.
  • domain assumption RoMa dense matching provides reliable correspondences in indoor scenes
    The inter-image depth prior and reprojection loss (Eqs. 7, 9, 14, 15) all depend on matched pixel pairs from the pretrained RoMa network; no analysis of match failure cases is given beyond the two filters.
  • domain assumption Monocular scale ambiguity is the primary failure cause in sparse-view reconstruction
    Motivates replacing monocular depth with triangulated depth; if scale ambiguity were not the main issue, the comparative advantage would not hold.
  • standard math Surface can be represented as zero-level set of an SDF optimized with VolSDF volume rendering
    Baseline representation adopted from Yariv et al. 2021, used throughout the method.
  • standard math Triangulation and epipolar geometry formulas are valid under pinhole camera model
    Used in Eqs. 8, 10, and 13 to compute depth priors, angular filters, and Sampson distances.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Sparis: Neural Implicit Surface Reconstruction of Indoor Scenes from Sparse Views." pith.science (2026). https://pith.science/paper/Z5QHIECL

@misc{pith2026250101196,
  author       = {Pith},
  title        = {Pith review of: Sparis: Neural Implicit Surface Reconstruction of Indoor Scenes from Sparse Views},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/Z5QHIECL}},
  note         = {Machine review of arXiv:2501.01196}
}
read the original abstract

In recent years, reconstructing indoor scene geometry from multi-view images has achieved encouraging accomplishments. Current methods incorporate monocular priors into neural implicit surface models to achieve high-quality reconstructions. However, these methods require hundreds of images for scene reconstruction. When only a limited number of views are available as input, the performance of monocular priors deteriorates due to scale ambiguity, leading to the collapse of the reconstructed scene geometry. In this paper, we propose a new method, named Sparis, for indoor surface reconstruction from sparse views. Specifically, we investigate the impact of monocular priors on sparse scene reconstruction, introducing a novel prior based on inter-image matching information. Our prior offers more accurate depth information while ensuring cross-view matching consistency. Additionally, we employ an angular filter strategy and an epipolar matching weight function, aiming to reduce errors due to view matching inaccuracies, thereby refining the inter-image prior for improved reconstruction accuracy. The experiments conducted on widely used benchmarks demonstrate superior performance in sparse-view scene reconstruction.

Figures

Figures reproduced from arXiv: 2501.01196 by the authors.

Figure 1
Figure 1. Surface reconstruction results from sparse views of an indoor scene. Our method Sparis outperforms in addressing [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. The overview of Sparis. Given sparse indoor images, the reconstruction of 3D surfaces is achieved via a 2-stage [PITH_FULL_IMAGE:figures/full_fig_p003_2.png] view at source ↗
Figure 3
Figure 3. Illustration of matching priors. (a) Using matching [PITH_FULL_IMAGE:figures/full_fig_p004_3.png] view at source ↗
Figures from the paper (9 more)
Figure 4
Figure 4. Figure 4: Visual comparisons of 3D reconstruction results on ScanNet with sparse views. The overall top views and the zoom-in [PITH_FULL_IMAGE:figures/full_fig_p005_4.png]
Figure 5
Figure 5. Figure 5: Visual comparisons of 3D reconstruction results [PITH_FULL_IMAGE:figures/full_fig_p006_5.png]
Figure 6
Figure 6. Figure 6: Our reconstruction results with or without match [PITH_FULL_IMAGE:figures/full_fig_p007_6.png]
Figure 1
Figure 1. Figure 1: Network architecture of our model. γ1(·), γ2(·) denote Positional Encoding and View Direction Encoding respectively. The image code is encoded from the sample batch using a linear layer. The surface normal n = ∇fi(x). In this supplementary document, we provide addition…
Figure 2
Figure 2. Figure 2: Visual comparison of reconstruction results with different pose settings on ScanNet with sparse input views. [PITH_FULL_IMAGE:figures/full_fig_p012_2.png]
Figure 3
Figure 3. Figure 3: Visual comparisons of GT poses and COLMAP [PITH_FULL_IMAGE:figures/full_fig_p012_3.png]
Figure 4
Figure 4. Figure 4: Visual comparisons of room-scale surface reconstruction results on scene0009 [PITH_FULL_IMAGE:figures/full_fig_p015_4.png]
Figure 5
Figure 5. Figure 5: Visual comparisons of room-scale surface reconstruction results on scene0710 [PITH_FULL_IMAGE:figures/full_fig_p016_5.png]
Figure 6
Figure 6. Figure 6: Visual comparisons of room-scale surface reconstruction results on room0, room1, room2, office0, office1, office2, [PITH_FULL_IMAGE:figures/full_fig_p017_6.png]

Discussion (0). Continue with ORCID to comment.

Reference graph

Works this paper leans on

50 extracted references · 25 canonical work pages

  1. [1]

    T.; Mildenhall, B.; Tancik, M.; Hedman, P.; Martin-Brualla, R.; and Srinivasan, P

    Barron, J. T.; Mildenhall, B.; Tancik, M.; Hedman, P.; Martin-Brualla, R.; and Srinivasan, P. P. 2021. Mip-nerf: A multiscale representation for anti-aliasing neural radiance fields. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 5855--5864

    2021

  2. [2]

    T.; Mildenhall, B.; Verbin, D.; Srinivasan, P

    Barron, J. T.; Mildenhall, B.; Verbin, D.; Srinivasan, P. P.; and Hedman, P. 2022. Mip-nerf 360: Unbounded anti-aliased neural radiance fields. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5470--5479

    2022

  3. [3]

    Chen, A.; Xu, Z.; Zhao, F.; Zhang, X.; Xiang, F.; Yu, J.; and Su, H. 2021. Mvsnerf: Fast generalizable radiance field reconstruction from multi-view stereo. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 14124--14133

    2021

  4. [4]

    Cong, W.; Liang, H.; Wang, P.; Fan, Z.; Chen, T.; Varma, M.; Wang, Y.; and Wang, Z. 2023. Enhancing nerf akin to enhancing llms: Generalizable nerf transformer with mixture-of-view-experts. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 3193--3204

    work page 2023

  5. [5]

    X.; Savva, M.; Halber, M.; Funkhouser, T.; and Nie ner, M

    Dai, A.; Chang, A. X.; Savva, M.; Halber, M.; Funkhouser, T.; and Nie ner, M. 2017. Scannet: Richly-annotated 3d reconstructions of indoor scenes. In Proceedings of the IEEE conference on computer vision and pattern recognition, 5828--5839

    2017

  6. [6]

    Ding, Y.; Yuan, W.; Zhu, Q.; Zhang, H.; Liu, X.; Wang, Y.; and Liu, X. 2022. Transmvsnet: Global context-aware multi-view stereo network with transformers. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, 8585--8594

    work page 2022

  7. [7]

    o kman, G.; Wadenb \

    Edstedt, J.; Sun, Q.; B \"o kman, G.; Wadenb \"a ck, M.; and Felsberg, M. 2023. RoMa: Revisiting Robust Losses for Dense Feature Matching. arXiv preprint arXiv:2305.15404

    arXiv 2023

  8. [8]

    Eftekhar, A.; Sax, A.; Malik, J.; and Zamir, A. 2021. Omnidata: A Scalable Pipeline for Making Multi-Task Mid-Level Vision Datasets From 3D Scans. In Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 10786--10796

    work page 2021

Show all 50 references
  1. [9]

    Gao, Y.; Cao, Y.-P.; and Shan, Y. 2023. SurfelNeRF: Neural Surfel Radiance Fields for Online Photorealistic Reconstruction of Indoor Scenes. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), 108--118

    2023

  2. [10]

    Gropp, A.; Yariv, L.; Haim, N.; Atzmon, M.; and Lipman, Y. 2020. Implicit geometric regularization for learning shapes. arXiv preprint arXiv:2002.10099

    2020 arXiv

  3. [11]

    Guo, H.; Peng, S.; Lin, H.; Wang, Q.; Zhang, G.; Bao, H.; and Zhou, X. 2022. Neural 3d scene reconstruction with the manhattan-world assumption. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5511--5520

    2022

  4. [12]

    Han, L.; Zhou, J.; Liu, Y.-S.; and Han, Z. 2024. Binocular-Guided 3D Gaussian Splatting with View Consistency for Sparse View Synthesis. In Advances in Neural Information Processing Systems (NeurIPS)

    2024

  5. [13]

    Huang, H.; Wu, Y.; Zhou, J.; Gao, G.; Gu, M.; and Liu, Y. 2023. NeuSurf: On-Surface Priors for Neural Surface Reconstruction from Sparse Input Views. arXiv preprint arXiv:2312.13977

    2023 arXiv

  6. [14]

    Z.; Zakharov, S.; Liu, K.; Guizilini, V.; Kollar, T.; Gaidon, A.; Kira, Z.; and Ambrus, R

    Irshad, M. Z.; Zakharov, S.; Liu, K.; Guizilini, V.; Kollar, T.; Gaidon, A.; Kira, Z.; and Ambrus, R. 2023. Neo 360: Neural fields for sparse view synthesis of outdoor scenes. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 9187--9198

    2023

  7. [15]

    M.; Lepoittevin, Y.; and Fleuret, F

    Johari, M. M.; Lepoittevin, Y.; and Fleuret, F. 2022. Geonerf: Generalizing nerf with geometry priors. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 18365--18375

    2022

  8. [16]

    Kazhdan, M.; Bolitho, M.; and Hoppe, H. 2006. Poisson surface reconstruction. In Proceedings of the fourth Eurographics symposium on Geometry processing, volume 7, 0

    2006

  9. [17]

    Levy, D.; Peleg, A.; Pearl, N.; Rosenbaum, D.; Akkaynak, D.; Korman, S.; and Treibitz, T. 2023. SeaThru-NeRF: Neural Radiance Fields in Scattering Media. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 56--65

    2023

  10. [18]

    Liang, Z.; Huang, Z.; Ding, C.; and Jia, K. 2023. HelixSurf: A Robust and Efficient Neural Implicit Surface Learning of Indoor Scenes with Iterative Intertwined Regularization. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 13165--13174

    2023

  11. [19]

    Long, X.; Lin, C.; Wang, P.; Komura, T.; and Wang, W. 2022. Sparseneus: Fast generalizable neural surface reconstruction from sparse views. In European Conference on Computer Vision, 210--227. Springer

    2022

  12. [20]

    Mar \' , R.; Facciolo, G.; and Ehret, T. 2022. Sat-nerf: Learning multi-view satellite photogrammetry with transient objects and shadow modeling using rpc cameras. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 1311--1321

    2022

  13. [21]

    P.; Tancik, M.; Barron, J

    Mildenhall, B.; Srinivasan, P. P.; Tancik, M.; Barron, J. T.; Ramamoorthi, R.; and Ng, R. 2020. NeRF: Representing Scenes as Neural Radiance Fields for View Synthesis. In Vedaldi, A.; Bischof, H.; Brox, T.; and Frahm, J.-M., eds., Computer Vision -- ECCV 2020, 405--421. Cham: ...

    2020

  14. [22]

    M\" u ller, T.; Evans, A.; Schied, C.; and Keller, A. 2022. Instant neural graphics primitives with a multiresolution hash encoding. ACM Trans. Graph., 41(4)

    2022

  15. [23]

    Reiser, C.; Peng, S.; Liao, Y.; and Geiger, A. 2021. Kilonerf: Speeding up neural radiance fields with thousands of tiny mlps. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 14335--14345

    2021

  16. [24]

    P.; Barron, J

    Rematas, K.; Liu, A.; Srinivasan, P. P.; Barron, J. T.; Tagliasacchi, A.; Funkhouser, T.; and Ferrari, V. 2022. Urban radiance fields. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12932--12942

    2022

  17. [25]

    Ren, Y.; Zhang, T.; Pollefeys, M.; S \"u sstrunk, S.; and Wang, F. 2023. Volrecon: Volume rendering of signed ray distance functions for generalizable multi-view reconstruction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 16685--16695

    2023

  18. [26]

    T.; Mildenhall, B.; Srinivasan, P

    Roessle, B.; Barron, J. T.; Mildenhall, B.; Srinivasan, P. P.; and Nie ner, M. 2022. Dense depth priors for neural radiance fields from sparse input views. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12892--12901

    2022

  19. [27]

    L.; and Frahm, J.-M

    Schonberger, J. L.; and Frahm, J.-M. 2016. Structure-from-motion revisited. In Proceedings of the IEEE conference on computer vision and pattern recognition, 4104--4113

    2016

  20. [28]

    Song, J.; Park, S.; An, H.; Cho, S.; Kwak, M.-S.; Cho, S.; and Kim, S. 2023. DäRF: Boosting Radiance Fields from Sparse Inputs with Monocular Depth Adaptation. arXiv:2305.19201

    2023 arXiv

  21. [29]

    J.; Mur-Artal, R.; Ren, C.; Verma, S.; et al

    Straub, J.; Whelan, T.; Ma, L.; Chen, Y.; Wijmans, E.; Green, S.; Engel, J. J.; Mur-Artal, R.; Ren, C.; Verma, S.; et al. 2019. The Replica dataset: A digital replica of indoor spaces. arXiv preprint arXiv:1906.05797

    2019 arXiv

  22. [30]

    Sun, C.; Sun, M.; and Chen, H.-T. 2022. Direct voxel grid optimization: Super-fast convergence for radiance fields reconstruction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 5459--5469

    2022

  23. [31]

    P.; Barron, J

    Tancik, M.; Casser, V.; Yan, X.; Pradhan, S.; Mildenhall, B.; Srinivasan, P. P.; Barron, J. T.; and Kretzschmar, H. 2022. Block-nerf: Scalable large scene neural view synthesis. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8248--8258

    2022

  24. [32]

    Turki, H.; Ramanan, D.; and Satyanarayanan, M. 2022. Mega-nerf: Scalable construction of large-scale nerfs for virtual fly-throughs. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 12922--12931

    2022

  25. [33]

    A.; Martin-Brualla, R.; Guibas, L.; and Li, K

    Uy, M. A.; Martin-Brualla, R.; Guibas, L.; and Li, K. 2023. SCADE: NeRFs from Space Carving with Ambiguity-Aware Depth Estimates. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 16518--16527

    2023

  26. [34]

    Wang, J.; Wang, P.; Long, X.; Theobalt, C.; Komura, T.; Liu, L.; and Wang, W. 2022 a . Neuris: Neural reconstruction of indoor scenes using normal priors. In European Conference on Computer Vision, 139--155. Springer

    2022

  27. [35]

    Wang, P.; Liu, L.; Liu, Y.; Theobalt, C.; Komura, T.; and Wang, W. 2021. Neus: Learning neural implicit surfaces by volume rendering for multi-view reconstruction. arXiv preprint arXiv:2106.10689

    2021 arXiv

  28. [36]

    Wang, X.; Dong, S.; Zheng, Y.; and Yang, Y. 2024. InfoNorm: Mutual Information Shaping of Normals for Sparse-View Reconstruction. arXiv preprint arXiv:2407.12661

    2024 arXiv

  29. [37]

    Wang, Y.; Han, Q.; Habermann, M.; Daniilidis, K.; Theobalt, C.; and Liu, L. 2023. Neus2: Fast learning of neural implicit surfaces for multi-view reconstruction. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 3295--3306

    2023

  30. [38]

    Wang, Y.; Li, Y.; Liu, P.; Dai, T.; and Xia, S.-T. 2022 b . NeXT: Towards High Quality Neural Radiance Fields via Multi-skip Transformer. In European Conference on Computer Vision, 69--86. Springer

    2022

  31. [39]

    Wu, H.; Graikos, A.; and Samaras, D. 2023. S-VolSDF: Sparse Multi-View Stereo Regularization of Neural Implicit Surfaces. arXiv preprint arXiv:2303.17712

    2023 arXiv

  32. [40]

    Xu, L.; Guan, T.; Wang, Y.; Liu, W.; Zeng, Z.; Wang, J.; and Yang, W. 2023. C2F2NeUS: Cascade Cost Frustum Fusion for High Fidelity and Generalizable Neural Surface Reconstruction. arXiv preprint arXiv:2306.10003

    2023 arXiv

  33. [41]

    Yao, Y.; Luo, Z.; Li, S.; Fang, T.; and Quan, L. 2018. Mvsnet: Depth inference for unstructured multi-view stereo. In Proceedings of the European conference on computer vision (ECCV), 767--783

    2018

  34. [42]

    Yariv, L.; Gu, J.; Kasten, Y.; and Lipman, Y. 2021. Volume rendering of neural implicit surfaces. Advances in Neural Information Processing Systems, 34: 4805--4815

    2021

  35. [43]

    Ye, B.; Liu, S.; Li, X.; and Yang, M.-H. 2023. Self-Supervised Super-Plane for Neural 3D Reconstruction. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 21415--21424

    2023

  36. [44]

    Ying, H.; Jiang, B.; Zhang, J.; Xu, D.; Yu, T.; Dai, Q.; and Fang, L. 2023. PARF: Primitive-Aware Radiance Fusion for Indoor Scene Novel View Synthesis. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 17706--17716

    2023

  37. [45]

    Yu, A.; Li, R.; Tancik, M.; Li, H.; Ng, R.; and Kanazawa, A. 2021. Plenoctrees for real-time rendering of neural radiance fields. In Proceedings of the IEEE/CVF International Conference on Computer Vision, 5752--5761

    2021

  38. [46]

    Yu, Z.; Peng, S.; Niemeyer, M.; Sattler, T.; and Geiger, A. 2022. Monosdf: Exploring monocular geometric cues for neural implicit surface reconstruction. Advances in neural information processing systems, 35: 25018--25032

    2022

  39. [47]

    Zhang, W.; Xing, R.; Zeng, Y.; Liu, Y.-S.; Shi, K.; and Han, Z. 2023 a . Fast Learning Radiance Fields by Shooting Much Fewer Rays. IEEE Transactions on Image Processing, 32: 2703--2718

    2023

  40. [48]

    Zhang, X.; Kundu, A.; Funkhouser, T.; Guibas, L.; Su, H.; and Genova, K. 2023 b . Nerflets: Local radiance fields for efficient structure-aware 3d scene representation from 2d supervision. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, 8274--8284

    2023

  41. [49]

    , " * write output.state after.block = add.period write newline

    ENTRY address archivePrefix author booktitle chapter edition editor eid eprint howpublished institution isbn journal key month note number organization pages publisher school series title type volume year label extra.label sort.label short.list INTEGERS output.state before.all...

  42. [50]

    write newline

    " write newline "" before.all 'output.state := FUNCTION n.dashify 't := "" t empty not t #1 #1 substring "-" = t #1 #2 substring "--" = not "--" * t #2 global.max substring 't := t #1 #1 substring "-" = "-" * t #2 global.max substring 't := while if t #1 #1 substring * t #2 gl...

Pith tools

Reviewed August 10, 2026 · model on record in the stance chip above.